Project Summary Cancer remains a leading cause of mortality among older adults in the United States. However, despite established guidelines supporting appropriate cancer screening in order to prevent death and adverse cancer- related outcomes, screening is often underutilized. Barriers to appropriate screening are even greater for groups facing long-standing screening disparities, such as racial/ethnic minorities and individuals with low socioeconomic status (SES). Key drivers of screening underutilization are decision-making biases facing clinicians and patients, highlighting the need for scalable solutions that are designed to address these biases and tailored to overcome barriers facing high-risk patients. Nudges, interventions designed using behavioral economic principles, improve behavior by addressing biases that lead to suboptimal decisions. Our team has extensive experience working with health systems to test and scale nudge interventions to overcome decision-making biases facing patients and/or clinicians. We have also demonstrated how these types of nudges can improve health care value and patient outcomes. Since electronic health records (EHRs) have been adopted by more than 90% of clinicians in the US, it is an ideal platform upon which to deploy large scale behavior change nudge interventions. In this study, we propose to personalize nudge interventions to clinicians and patients, with a focus on how to tailor nudges to the needs of high-risk patients and how to implement promising nudges to improve cancer screening among older adults. We will pilot this work at Penn Medicine and then implement a pragmatic trial at sites in the Penn Medicine, University Hospitals, and Sutter Health systems. In the R61 phase, we will complete the following aims at Penn Medicine: analyze EHR and claims data and identify characteristics and subgroups of patients at high-risk for not completing cancer screening (Aim 1), test the feasibility of individual nudge interventions within an EHR nudge toolkit (Aim 2), and pilot test nudge interventions to identify the most promising approaches to increase cancer screening rates for high-risk patients in each subgroup (Aim 3). In the R33 phase, we will achieve the following aims at Penn Medicine, University Hospitals, and Sutter Health: conduct a two-arm pragmatic cluster-randomized controlled trial to test the effectiveness of personalized EHR-based nudges to clinicians and patients on increasing cancer screening rates (Aim 1), evaluate if EHR-based nudge interventions reduce disparities in cancer screening rates for racial/ethnic minorities and patients with low socioeconomic status (Aim 2), and inform future design of tailored EHR interventions by examining heterogeneity of treatment response with respect to clinician and patient characteristics (Aim 3).